INDEX

Explanations

expressions of gratitude and positive affirmations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Resource Center

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

MigrationBuilder

-0.56

 يتيمه

-0.52

 Comprometido

-0.51

 nahilalakip

-0.50

出版年

-0.50

 Bewußt

-0.50

WithIOException

-0.49

LEncoder

-0.49

 autorytatywna

-0.48

 correctes

-0.46

POSITIVE LOGITS

 Lastly

0.38

 lại

0.37

 Overall

0.36

glomer

0.35

 Ultra

0.35

 lastly

0.35

 overall

0.35

 leger

0.35

cil

0.34

 Inform

0.34

Activations Density 0.006%

expressions of gratitude and positive affirmations

Resource Center

No Comments

No Known Activations